Applied Time Series Analysis

This course considers statistical techniques to evaluate processes occurring through time. It introduces students to time series methods and the applications of these methods to different types of data in various contexts (such as actuarial studies, climatology, economics, finance, geography, meteorology, political science, risk management, and sociology). Time series modelling techniques will be considered with reference to their use in forecasting where suitable. While linear models will be examined in some detail, extensions to non-linear models will also be considered. The topics will include: deterministic models; linear time series models, stationary models, homogeneous non-stationary models; the Box-Jenkins approach; intervention models; non-linear models; time-series regression; time-series smoothing; case studies. Statistical software R will be used throughout this course. Heavy emphasis will be given to fundamental concepts and applied work. Since this is a course on applying time series techniques, different examples will be considered whenever appropriate.